1. Field of the Invention
The present invention relates to a circuit  device and method for controlling recording and reproduction in a digital cassette tape  recorder, and to an apparatus and method for recording or reproducing digital data to or from a digital storage medium using speed information.
2. Description of the Background Art
For analog/digital conversion for converting an analog video signal into a digital video signal and linear quantization, a signal transmission rate of about 100 Mega bits per second is typically required in the case of a normal TV broadcast signal such as NTSC, SECAM and PAL signals. On the other hand, a high definition TV (HDTV) signal with higher resolution than that of the normal TV broadcast signal requires a signal transmission rate higher than 100 Mega bits.
For achieving data transmission in a limited transmission band, digitized video signals should be transmitted in the form compressed in accordance with the video data compression-technique. In the case of digital cassette tape recorders (digital VCRs) having a limitation on record bandwidth, signals recorded on a magnetic tape may be digital normal TV signals having the form of compressed signals or digital HDTV signals having the form of compressed signals.
Referring to FIG. 1, there is illustrated a conventional recording circuit for such  a digital VCR  video recorder. Here, a digital VCR as one example of a digital video recorder is discussed. As shown in FIG. 1, the recording circuit includes an interface 1 for converting a compressed digital video signal into a signal having the recordable form, an interleaving and channel-dividing circuit 2 for interleaving an output V1 from the interface 1 in a predetermined form to reduce burst error and channel-dividing it to be matched with the zero bandwidth of a recording channel, recording formatters 3A and 3B for respectively converting outputs V2 and V3 of the interleaving and channel-dividing circuit 2 to record formats each including a synchronous signal, an identification signal and redundancy bits for error correction codes, channel modulators 4A and 4B for modulating outputs V4 and V5 of the recording formatters 3A and 3B, respectively, recording amplifiers 5A and 5B for amplifying outputs V6 and V7 of the channel modulators 4A and 4B, respectively, a drum pulse generator 7 for outputting two pulses at every rotation of a head drum 6 caused by driving a drum motor M1, and switches SW1 and SW2 for performing their switching operations based on an output SWP from the drum pulse generator 7 to selectively transmit outputs V8 and V9 of the recording amplifiers 5A and 5B to heads HD1 (or HD3) and HD2 (or HD4), respectively. In FIG. 1, the reference numeral 8 denotes a guide pin, 10 a pinch roller, and 9 a capstan adapted to be rotated by a capstan motor M2.
FIG. 2 is a block diagram illustrating a conventional reproduction circuit for the digital VCR  recorder. As shown in FIG. 2, the reproduction circuit includes reproduction amplifiers 11A and 11B for receiving outputs from selected heads HD1 (or HD3) and HD2 (or HD4) mounted on the head drum 6 via the switches SW1 and SW2 switched in accordance with the output SWP from the drum pulse generator 7 and amplifying them, respectively, equalizers 12A and 12B for compensating distortions of frequency characteristics of outputs V10 and V11 of the reproduction amplifiers 11A and 11B, respectively, channel demodulators 13A and 13B for demodulating outputs V12 and V13 of the equalizers 12A and 12B, respectively, sync-detecting and error-correcting circuits 14A and 14B for detecting synchronous signals SYNC added in a recorded signal from the outputs V14 and V15 of the channel demodulators 13A and 13B and correcting errors of the outputs V14 and V15, respectively, deinterleaving circuits 15A and 15B for deinterleaving outputs V16 and V17 of the sync-detecting and error-correcting circuits 14A and 14B into the original signal form, respectively, a deformatter 16 for recovering outputs V18 and V19 of the deinterleaving circuits 15A and 15B to the original signal format, and an interface 17 for converting an output V20 of the deformatter 16 into a reproduced digital signal Vo and outputting it.
Now, operations of the conventional circuits will be described in conjunction with FIGS. 1 to 6.
First, in a recording mode, a compressed HDTV signal or compressed normal TV signal is applied to the interface 1 which, in turn, converts the received signal into a signal V1 capable of being recorded and reproduced. The signal V1 is then inverted into a predetermined form to reduce burst errors in the interleaving and channel-driving circuit 2 which, in turn, outputs signals V2 and V3 channel-divided so as to be matched with the bandwidth of the recording channel.
The outputs V2 and V3 from the interleaving and channel-dividing circuit 2 are applied to the recording formatters 3A and 3B and then added with synchronous signals SYNC, identification signals ID and redundancy bits for error correction codes ECC in the recording formatters 3A and 3B. Resultant signals from the recording formatters 3A and 3B are then received in the channel modulators 4A and 4B which, in turn, output signals V6 and V7 matched with a predetermined recording format, respectively. The outputs V6 and V7 from the channel modulators 4A and 4B are applied to the recording amplifiers 5A and 5B which, in turn, amplify them, respectively.
Outputs V8 and V9 from the recording amplifiers 5A and 5B are applied to selected heads HD1 (or HD3) and HD2 (or HD4) via the switches SW1 and SW2 switched by the output SWP from the drum pulse generator 7, so that they are recorded on a magnetic tape  recording medium in a recording format shown in FIG 3.
In this case, the drum pulse generator 7 generates two pulses at every rotation of the head drum 6 driven by the drum motor M1.
Meanwhile, frames have a mixed form of intra-frames (I-frames) able to be independently decoded and predictive frames (P-frames) compressed by moving information of previous screen and unable to be independently decoded, in accordance with a video compression system for HDTV signals or an MPEG (Moving Picture Experts Group) system. Bit rate generated in each frame is non-uniform, as shown in FIG. 5.
In a reproduction mode, the magnetic tape travels by the rotation of the capstan 9 caused by the capstan motor M2 while being in contact with the head drum 6 rotating by the driving force of the drum motor M1. At this time, the heads HD1 (or HD3) and HD2 (or HD4) detect signals on the magnetic tape and send them to the reproduction amplifiers 11A and 11B via the switches SW3 and SW4 switched by the output SWP of the drum pulse generator 7, respectively.
The signals received in the reproduction amplifiers 11A and 11B are amplified to a predetermined level and then sent to the equalizers 12A and 12B which, in turn, output signals V12 and V13 having compensated frequency characteristics, respectively. The signals V12 and V13 from the equalizers 12A and 12B are then applied to the channel demodulators 13A and 13B, respectively, so as to be demodulated. Outputs V14 and V15 from the channel demodulators 13A and 13B are received in the sync-detecting and error-correcting circuits 14A and 14B which, in turn, detect respectively synchronous signals SYNC and identification signals ID from synchronous block of the received signals and remove error components included in the data.
Outputs V16 and V17 from the sync-detecting and error-correcting circuits 14A and 14B are applied to the deinterleaving circuit 15A and 15B which, in turn, deinterleave the signals V16 and V17 and thereby generate signals V18 and V19 having the original signal forms, respectively. The signals V18 and V19 are received in the deformatter 16 and thereby converted to the format having the signal form prior to recording. Signal V20 from the deformatter 16 is applied to the interface 17 which, in turn, generates a reproduced digital signal Vo.
In a speed-varied reproduction, the rotation speed of the head drum 6 is kept constant while the travel of the magnetic tape is accelerated. As a result, the heads HD1 to HD4 travel across tracks on the magnetic tape. The trace of the heads is shown in FIG. 3. Consequently, the detected signals have a discontinuous data form, data burst with a magnitude inversely proportional to the travel speed of the magnetic tape.
In the case of existing analog VCRs  video recorders, data of one field are recorded in one track in the reproduction order. Accordingly, regions on tracks from which data are detected in the speed-varied reproduction mode are directly associated with reproduction regions of a corresponding screen. Therefore, video reproduction in the speed-varied mode is possible even when a noise bar is generated due to data detected on an adjacent track.
In the case of existing digital VCRs  video recorders, however, data of one field are recorded in a plurality of tracks, as shown in FIG. 3. As a result, reproduction bursts on adjacent tracks have no relation with the reproduction order. In this case, therefore, a frame memory and an addressing process for rearranging data detected are needed. Furthermore, there is a problem of an inevitable mosaic-shaped distortion of small segments due to discontinuous detection of data bursts.
For a video reproduction in the existing digital VCRs  video recorders, data bursts detected should be independently decoded. However, these data bursts include unrecoverable other data on the screen or unrecoverable previous screen data, because the data bursts have the digital form compressed using the correlation between signals that may be the important factor of adversely affecting the picture quality in reproduction. The unrecoverable data can not be decoded and thereby reproduced in the form of videos. In particular  Further, such a problem becomes more frequent in the case of data obtained from the  a video compression systemsuch as the , for example, variable length coding involving non-uniform data lengths.
In other words  Moreover, although data bursts detected from tracks on which data for the 0-th I-frame, the n-th I-frame, the 2n-th I-frame . . . are recorded can be constructed to a video, data bursts detected from tracks on which data of P-frames are recorded can not be constructed to a video.